The Moshin Knee Brace is designed with athletes—both current and former—who have torn their ACL in mind. It serves as a rehabilitation tool, bridging the gap between post-surgery recovery and full rehabilitation. However, its reusability means it could also be valuable after full healing, as ACL re-tear rates remain high, around 20-25%. This means that one in every four to five athletes could suffer a second ACL tear.

Current knee brace designs for ACL injuries face several challenges: they typically restrict angles, with users or patients gradually adjusting the angle restriction mechanism, making them bulky and lacking versatility. Additionally, many knee braces become obsolete once the patient has fully healed. Finally, most designs only address knee range of motion (ROM), without considering the posterior ligaments that stretch up to the hip joint. The core kinematics of the knee include flexion, extension, abduction, adduction, and internal and external rotation.

Knee Biomechanics and ACL Loading Insights

Research indicates that anterior cruciate ligament (ACL) loading decreases as knee flexion increases and diminishes further for a given anterior force. Comparative studies on knee kinematics between healthy and diseased knees reveal notable differences: reduced range of motion (ROM) during stair ascent and descent, greater knee abduction angles during weight-bearing activities, and increased knee energy absorption in ACL-injured patients. Additionally, the approximate pressure exerted on the knee during bending is estimated at 1.5 times the body weight. These biomechanical considerations informed the design process.

Through discussions with an orthopedist in Korea and an ex-athlete who experienced an ACL tear, I decided to incorporate a BOA lacing system. This proven, reliable mechanism—commonly used in snowboard boots and footwear—provides a secure and adjustable fit to stabilize the knee and assist in leg straightening. The design also allows for easy adjustments beneath clothing, ensuring that users can manage the brace discreetly without drawing attention to the injury.

This innovative system provides both security and comfort throughout the rehabilitation process.

For example, allowing for easy adjustment to tighten and stabilize the knee, while also enabling rapid release of tension, such as when the user needs to sit down.

PROBLEM STATEMENT

PROCESS

Incorporating Accelerometer Technology for Knee Safety

Traditional knee braces often overlook the tension of ligaments that extend to the hip. To address this, the design integrates an accelerometer sensor. This sensor, paired with a vibration motor, provides haptic feedback to alert users of potentially dangerous movements.
Accelerometers, commonly found in devices like smartphones, airplanes, and video game controllers, measure velocity changes and detect orientation and rotation in three dimensions. The decision to include an accelerometer sensor in the design was driven by its ability to provide haptic feedback when the knee reaches dangerous velocity thresholds.

Advancing User Interaction Through Gesture Recognition and Machine Learning

As technology evolves, so does the capability of computers to interpret user interaction with the physical world. Gesture recognition, a key feature of the system, ensures accurate differentiation between normal and irregular movements, further refining the user experience.

The accelerometer is integrated with an external machine learning platform, EdgeML, which simplifies data analysis to detect anomalies or irregularities in walking or running patterns. The system employs gesture recognition to filter out random, non-pattern movements (e.g., spasms) during the training phase, ensuring these are not misinterpreted as events. This adaptability enables the feedback mechanism to adjust across individual users and diverse knee movements, enhancing safety and functionality.

The knee brace design incorporates high-quality materials to ensure functionality and comfort. It features a North Face-manufactured mesh for durability, neoprene for rigidity (commonly used in knee braces), coated cotton for the exterior, and lycra edge binding. A discreet dial adjustment mechanism allows the brace to be easily adjusted underneath clothing, minimizing attention to the injury.

Personalized Monitoring and Feedback Through Machine Learning
To cater to individual needs and varying movement types, such as walking versus running, the accelerometer connects to an external machine learning platform, EdgeML. This program processes and simplifies extensive accelerometer data, identifying irregularities and transmitting this information to a microcontroller. Users can record their movement data (e.g., walking or running), adjust the vibration motor’s sensitivity threshold via a smartphone app, and track their knee movement and progress. This dual functionality enhances both safety and user confidence, emphasizing reassurance as much as actionable insights.